Debdatta Kandar
Papers
1
Total Citations
101
H-Index
1
About
Debdatta Kandar is a leading researcher in the intersection of deep learning and pattern recognition, with a primary focus on advancing Optical Character Recognition (OCR) and handwritten text analysis. His most impactful work, "Handwritten Character Recognition from Images using CNN-ECOC" (2020), has garnered 101 citations, introducing a novel hybrid architecture that combines Convolutional Neural Networks with Error-Correcting Output Codes to dramatically improve classification accuracy for complex, unconstrained handwritten scripts. This contribution addresses a critical bottleneck in digitizing historical documents and automating postal services. Beyond this landmark paper, Kandar’s research explores robust feature extraction and ensemble learning methods for image-based recognition, pushing the boundaries of how machines interpret human handwriting. His work is widely cited by engineers developing real-world OCR systems and by academics studying deep neural network optimization. Recognized for bridging theoretical advances with practical deployment, Kandar continues to influence the next generation of intelligent character recognition technologies, making him a key figure in applied computer vision and machine learning.
Research Focus
Key Achievements
Top Papers
- 1Handwritten Character Recognition from Images using CNN-ECOC101 citations · 2020